From the Department of Diagnostic and Interventional Neuroradiology, Klinikum Rechts der Isar, Technische Universität München.
Deepc GmbH, Munich, Germany.
Invest Radiol. 2021 Sep 1;56(9):571-578. doi: 10.1097/RLI.0000000000000775.
Anomaly detection systems can potentially uncover the entire spectrum of pathologies through deviations from a learned norm, meaningfully supporting the radiologist's workflow. We aim to report on the utility of a weakly supervised machine learning (ML) tool to detect pathologies in head computed tomography (CT) and adequately triage patients in an unselected patient cohort.
All patients having undergone a head CT at a tertiary care hospital in March 2020 were eligible for retrospective analysis. Only the first scan of each patient was included. Anomaly detection was performed using a weakly supervised ML technique. Anomalous findings were displayed on voxel-level and pooled to an anomaly score ranging from 0 to 1. Thresholds for this score classified patients into the 3 classes: "normal," "pathological," or "inconclusive." Expert-validated radiological reports with multiclass pathology labels were considered as ground truth. Test assessment was performed with receiver operator characteristics analysis; inconclusive results were pooled to "pathological" predictions for accuracy measurements. External validity was tested in a publicly available external data set (CQ500).
During the investigation period, 297 patients were referred for head CT of which 248 could be included. Definite ratings into normal/pathological were feasible in 167 patients (67.3%); 81 scans (32.7%) remained inconclusive. The area under the curve to differentiate normal from pathological scans was 0.95 (95% confidence interval, 0.92-0.98) for the study data set and 0.87 (95% confidence interval, 0.81-0.94) in external validation. The negative predictive value to exclude pathology if a scan was classified as "normal" was 100% (25/25), and the positive predictive value was 97.6% (137/141). Sensitivity and specificity were 100% and 86%, respectively. In patients with inconclusive ratings, pathologies were found in 26 (63%) of 41 cases.
Our study provides the first clinical evaluation of a weakly supervised anomaly detection system for brain imaging. In an unselected, consecutive patient cohort, definite classification into normal/diseased was feasible in approximately two thirds of scans, going along with an excellent diagnostic accuracy and perfect negative predictive value for excluding pathology. Moreover, anomaly heat maps provide important guidance toward pathology interpretation, also in cases with inconclusive ratings.
异常检测系统通过偏离学习到的规范，有可能发现整个病理谱，从而为放射科医生的工作流程提供有意义的支持。我们旨在报告一种弱监督机器学习（ML）工具在检测头 CT 中的病理方面的效用，并在未经选择的患者队列中充分对患者进行分诊。
在 2020 年 3 月，所有在三级保健医院接受头部 CT 的患者都有资格进行回顾性分析。仅纳入每位患者的第一次扫描。使用弱监督 ML 技术进行异常检测。异常发现以体素级显示，并汇总为 0 到 1 的异常评分。该评分的阈值将患者分为 3 类：“正常”、“病理”或“不确定”。具有多类病理标签的专家验证的放射学报告被认为是真实数据。使用接收器操作特征分析进行测试评估；不确定结果被汇总为“病理”预测，以进行准确性测量。外部有效性在公开的外部数据集（CQ500）中进行了测试。
在调查期间，有 297 名患者被转诊进行头部 CT，其中 248 名患者可纳入研究。在 167 名患者中（67.3%）可以明确地将评分归类为正常/病理；81 次扫描（32.7%）仍不确定。研究数据集区分正常和病理扫描的曲线下面积为 0.95（95%置信区间，0.92-0.98），外部验证的曲线下面积为 0.87（95%置信区间，0.81-0.94）。如果将扫描分类为“正常”，则排除病理的阴性预测值为 100%（25/25），阳性预测值为 97.6%（137/141）。敏感性和特异性分别为 100%和 86%。在评分不确定的患者中，41 例中有 26 例（63%）发现了病理。
本研究首次对脑部成像的弱监督异常检测系统进行了临床评估。在未选择的连续患者队列中，大约三分之二的扫描可以明确地分为正常/患病，同时具有出色的诊断准确性和排除病理的完美阴性预测值。此外，异常热图为病理解释提供了重要指导，即使在评分不确定的情况下也是如此。
